Metabolomics is an emerging discipline in the field of postgenomic research which allows to analyze hundreds or thousands of low molecular weight metabolites in intact tissues and/or biofluids such as urine, serum, plasma, saliva and so on. CERM has established a metabolomic laboratory called CEMET, with a dedicated NMR instrument and a small group of young and enthusiastic researchers.
The whole of metabolites is a very accurate and complex index of the health state of a human organism. One of the most suitable analytical techniques for this kind of studies is the Nuclear Magnetic Resonance (NMR) spectroscopy that permits to analyze at the same time all metabolites present in a biofluid with a minimum of sample pre-treatment. The investigation of this “finger-print” (by suitable statistical tools for multivariate data analysis) is a powerful mean to study the correlation between various pathologies and the metabolites and their relative variations. Being one of the ultimate aims of metabolomic studies the discrimination between “normal” and “non normal” subjects, that is finding differences between healthy and ill or potentially at risk subjects, we have carried out a NMR spectra database of urine sample of adult (about 1000 samples) subjects. Moreover, we are studying many plasma and serum samples from varied areas of interest in the biomedical field. The primary current metabolomic projects, are:

metabolomics_2Graphical representation of a “Natural” gender discrimination between 1H-NMR spectra of urine samples collected from men (blue) and women.

  • Potential Celiac disease (MetCPot)
  • Study of the progression of epatocarcinomas (HepaMet)
  • Broncho-pulmonary obstructive diseases (MetPol)
  • Study on metabolic fingerprint of cardiovascular risk (MetAVIS)
  • Tumor progression in geriatric patients (MetGer)
  • New statistical methods for data analysis and clustering
  • Prognostic studies on patients with metastatic colorectal cancer (MetCRC)
  • Analysis of saliva samples for periodontal diseases fingerprinting (MetSAL)
  • Studies on inflammatory bowel diseases (MetIDB)
  • The use of metabolomics for samples quality assessment
  • Discovery of new biomarker for diagnosis and prognosis of melanoma (MetMEL)

The study of these fluids is performed through nuclear magnetic resonance (NMR) spectroscopy. Data obtained from NMR spectra are analysed with advanced computational techniques such as supervisioned learning methods (Support Vector Machines, Neural Networks) and with standard techniques such as principal component analyses or multilinear regression methods.

Exampleof the variationinconcentration of12metabolitesin22 healthysubjects.